This invention relates to utilizing heat transfer equations to trade profitably upon the poor conductivity of porous uncured elastomers and the high conductivity of a metal mold, usually aluminum, in which a thick-walled article of a curable elastomer is cured under high pressure exerted by a hot fluid. By "thick-walled" I refer to a cross-section of elastomer, usually rubber, of sufficient thickness to provide a substantial heat-sink. For example, a green (uncured) tire carcass is cured (vulcanized) in a mold by heating with a fluid-pressurized bladder inserted within the tire, while the outer surface of the tire is heated by contact with the surfaces of a heated mold, usually heated with steam.
More specifically, the invention relates to a process for delivering the optimum number of cure equivalents to an article to be cured, this number of cure equivalents being delivered to the point of least cure (PLC), after the hot pressurized liquid on one surface of the article (the inner surface, say) is replaced with a cold liquid under the same pressure, while the other surface (the outer) is still being heated. The replacement of hot liquid by cold liquid, referred to as "change-over", is effected before the article is removed from the mold, without even momentarily releasing the pressure, and without sensing the temperature at any point within the article while it is being cured.
A "cure equivalent" is defined as one minute of curing time at a constant reference temperature, usually 280.degree. F. The PLC is so referred to because it is the critical point at which the desired number of cure equivalents is to be delivered. When neither more nor less cure equivalents than optimum are delivered to the PLC, the article is said to have had a "perfect cure". Methods of computing the number of cure equivalents to be delivered, or determining the location of the PLC are known and are only incidental to the present invention.
Since the pressure within the bladder is to be maintained when the temperature is quickly dropped, it will be evident that the process of this invention can only be practiced with a hot liquid being circulated in the bladder, and not steam.
The article may be a tire, snubber for a shock absorber, cushioning blocks for rairoad siding blocks, expansion joints for decks and bridges, molded railroad crossing beds, or any article of arbitrary shape, provided its thickness is sufficient to benefit from the lag time for the transfer of heat through a highly conductive metal mold. This lag time is a function of the physical properties of uncured elastomer, of the mold, the temperature and pressure of the curing fluids, and the unsteady state heat transfer relationships which control the curing of the article. Further reference to the article will be made by specifying a tire, more specifically an automobile, truck or "off-the-road" (OTR) tire in which the carcass has substantial thickness and the road-contacting portion, including the tread, is usually at least about 0.50 inch (about 12 mm) thick.
It has long been recognized that a hot tire, freshly removed from a curing mold continues to cure while it is cooling, though the temperature at the PLC begins to fall as soon as the tire is removed from the mold. More correctly, a tire is typically over-cured in areas closest to the heat sources even when the perfect cure is achieved at the PLC, and continues to overcure while cooling. Conventionally, to avoid excessive overcuring of a carcass, a cured carcass is removed from a curing press, rapidly mounted upon a former, inflated and left to cool to room temperature before it is discharged onto a conveyor for sorting, storage and shipping. Excessive overcuring of the carcass is far more deleterious than that of the tread, and it is fortunate that the geometry of a tire is such that in my process, the extent of overcuring the tread due to continued heating after the change-over, is no greater than it is in a conventional curing process.
A particular utilization of lag time is made in the curing of a rubberized nylon cord carcass as taught in U.S. Pat. No. 3,718,721 to Gould et al where the mold heating means is rendered inoperative when a predetermined state of cure has been reached. As clearly pointed out in col 5, lines 44-49, a satisfactory state of cure is reached and opening of the press is initiated when the comparison of the measured temperature with the reference temperature of cure and the elapsed time of cure indicates that the state of cure is such that no porosity will develop in the tire upon release of pressure within the press.
Obviously, if the state of cure is such that the tire will "blow" if the pressure is released, the tire will be destroyed. A tire is said to "blow" when its state of cure is such that enough gases (air, and those generated by the vulcanization reaction) are trapped within the rubber to expand the body, often with too great a force to be contained because the rubber is not sufficiently cured. When sufficiently cured, even if the optimum number of cure equivalents is not as yet delivered to the PLC, most of the entrapped gases have escaped through vents in the mold, the matrix of rubber is substantially thoroughly reinforced by virtue of the crosslinking of polymer chains, and releasing the pressure does not produce significantly greater porosity than that of properly cured rubber.
Thus, Gould et al teach that when a predetermined overall change in the total state of cure has been achieved, the computer actuates a valve mechanism which controls the steam supply to shut it off. At such an instant, the computer actuates a mold-opening mechanism and the completed tire is removed (see col 5, lines 13-19). Obviously, if at the instant the mold-opening mechanism is actuated, the temperature at any point in the tire is above that at which the tire will blow, the tire will be destroyed. Even if the mold was not opened, the cold flood could only be initiated after the cure was past the point when porosity would develop when the pressure was released. In other words, the mold could be cold-flooded only after the risk of blowing the tire had abated. This risk was a necessary consequence of having to release pressure, if only instantaneously, to make the change from a steam supply to a cold flood. It was only upon completion of the change-over that a cold flood could be initiated within the press to cool the tire. In so doing, namely cold flooding the mold immediately upon negating the risk of blowing the tire, they found a way of profitably utilizing that portion of the lag time which allowed stored heat to be utilized to set the nylon cords and minimize distortion of the carcass.
It was essential that the point where porosity would develop be passed before the change-over from saturated steam to cold water, because the sudden change in temperature will necessarily produce too great a drop in the vapor pressure of the saturated steam. If, for example, saturated steam at 200 psig (1479 kPa) and 388.degree. F. (198.degree. C.) was replaced by water at 100.degree. F. (37.8.degree. C.) before the rubber was cured to a point where porosity would develop because the rubber was as yet uncured, the pressure drop would be so great that the tire would blow. This is because the steam in the bladder cannot be instantaneously displaced, and the bladder filled with cold water. In reality, it takes some time to displace the steam from the bladder, and to fill the bladder with water. This is not to say that the steam cannot be first displaced by hot water at the same temperature as the saturated steam, so that there is no drop in pressure, the hot water supply in turn, being then displaced by cold water, referred to as a "cold-flood".
As will presently be evident, a change-over from a hot supply to a cold flood without at least momentarily releasing pressure would not be possible without using a valving sequence described hereinbelow. Making the changeover while the carcass is as yet uncured, and in a state such that the tire will blow upon release of pressure within the press, profitably utilizes the maximum amount of lag time without making any direct measurement of temperature within the curing tire.
The means for tracking the PLC without sensing temperature within the tire is disclosed in U.S. Pat. No. 4,371,483 to Mattson. However, Mattson's interest was sharply focussed upon the problem of mimicing a three-dimensional finite difference program to track the PLC without sensing a temperature within the body. Ten years after Gould et al '721, it was up to Mattson '483 to provide an effective curing process without sensing temperature. The problem he solved was formidable enough without interjecting yet another variable, namely cold flooding one side of the mold, typically the bladder side, while the other side continued to be heated. There was no reason to read past the problems of defining the location of the PLC, and quantifying the number of cure equivalents delivered to that point, which problems were addressed in Gould et al and Mattson, to attempt to address the problem of switching from hot to cold flood substantially instantaneously, without releasing pressure, at a temperature at which the tire would otherwise blow.